Google this week launched, Allo, the latest in its efforts at social. We’ve seen a long Wave of Google social products that have failed. Buzz, Wave, OpenSocial, Google+ on the pure social side. When you look at the subset of messaging apps, this includes gTalk, Google Voice, Google Hangouts among others.
Allo is Google’s latest attempt to compete with Facebook Messenger, iMessage, WhatsApp and Skype.
There is no clear reason to adopt this. Why is a user going to adopt Allo? Is it for:
- Tons of emojis. (Piece of cake to emulate.)
- To play command line games? Zork 2016 (Piece of cake to emulate.)
- Google Assistant.
- whisper SHOUT. (Piece of cake to emulate. iOS 10 includes this.)
Better to pick one thing and knock that out of the ballpark. You aren’t going to win FB Messenger users over with emoji. Given Google and Facebook’s relative strengths and weaknesses, I’d bet it all on Google Assistant. Another plus: It adds virality to Google’s other products.
The initial implementation of the assistant is an OK start, but there’s a long, long way to go. Google Assistant is like most bots, it overpromises and underdelivers.
One of the challenges in natural language processing is understanding entities. When I asked a friend “Do you want to meet up at blue line pizza tonight?”, I got a search suggestion for “Pizza places nearby”. It didn’t recognize that “blue line pizza” is an actual place. When I said “How about tacorea?” It gave me the correct suggestion of “Tacorea restaurant”.
Having worked in local, search and messaging, I know that entity extraction is an incredibly hard technical problem. So I’m going to be more forgiving than most people. A lot of users will just feel that the experience is broken.
Google is also behind in another way: Unlike Facebook and iMessage (and even Google Hangouts), there is no desktop experience. I wanted to send a link to this post to a friend over Allo (after I wrote it on my Mac), but had to send it via Hangouts instead.
The biggest challenge for Allo will be distribution. I already have plenty of ways to message someone: Facebook Messenger, WhatsApp, Skype, SMS, Line, Twitter DM, iMessage.
iMessage succeeded because it Apple just took over SMS transport for iPhone to iPhone messaging. (Apple was able to do this because it has always been able to dictate the rules to carriers.)
WhatsApp built its base outside the U.S. The primary reason people adopted it initially was to avoid paying the exorbitant cross-border SMS and MMS fees. There was an easy, compelling reason to switch.
Facebook Messenger used its insane time-on-site and hundreds of millions of users to build its user base. They had a massive (and personal) friend graph to work with.
So far, I haven’t seen anything from Google about how it’s going to attract users.
If you’ve been reading this blog or following my tweets, you know that I’m a huge proponent of self-driving cars. In the long run, they will save lives, reduce environmental costs of transportation and make more efficient use of capital. They will fundamentally change the nature of cities and society.
But we’re not there yet. And we won’t be for many years to come.
A Tesla enthusiast died recently when his Model S drove straight into a truck that was making a turn. The car’s “Autopilot” mode didn’t recognize the brightly colored truck against the brightly colored sky. Neither did he. (A portable DVD player was found in his car; it isn’t clear if he was watching it. A witness said that it was playing Harry Potter shortly after the accident.)
We’re in the midst of a long transition period in cars and car safety. I’m afraid this won’t be the last such incident.
We have many different kinds of safety and driver-assistance features in cars today. Some assist driving. Others offer semi-automation. The last category is true autonomous vehicles. (There are no vehicles of the last type in commercial production.)
Definitions of what belongs what will vary. But this is how I think about them.
Driver assistant features
These help the driver with alerts or by managing small parts of the driving experience. They check the work of the driver. They include:
- Anti-lock brakes. The system pulses the brakes to help prevent skidding. Before anti-lock brakes, drivers had to manually pump brakes to keep from hard braking and locking the wheels. With ABS, the system pulses the brakes much faster than a human can. The braking has to be initiated by the driver. These are standard on U.S. cars.
- Back up cameras and back up sensors. When the car is put into reverse, back up sensors will beep as it detects an object behind you. The closer you get to the object, the more frequent the beeping. Cameras show you what’s behind the car, including things you wouldn’t see in the rearview mirror. Cameras are now in about 60% of new vehicles in U.S.; they will be required in cars by 2018.
- Lane-departure warning systems. These notify you when you are drifting out of your lane. They use cameras to look for lane markings. The driver still has to do the steering; the system only alerts to mistakes. LDWS are options on mid- to high-end cars.
- Blindspot detection. When you are changing lanes, blindspot detection systems will alert you when there is a vehicle in your blindspot. This could be an audible alert or an indicator in the side view mirror. BSDs are options on mid- to high-end cars.
These are typically offered on mid- to high-end cars. They actively control the vehicle. They include:
- Cruise control. Cruise allows a driver to set a steady speed for the vehicle and it will maintain the speed. The driver can then remove the foot from the accelerator. Even in light traffic, this is a pretty useless feature. Because other cars change speeds, you have to keep adjusting the cruise setting. This has been a common feature for decades.
- Adaptive cruise control. Similar to cruise control, but the speed adapts to the car in front of you. If the car slows down, your car will slow down.
- Lane management systems. They will keep you in your lane by using cameras to detect lane markings. They’re rarer than LDWS, but rely on the same basic technology.
- Automatic braking. These detect imminent collisions and automatically apply the brakes.
- Automatic parallel parking. These will park your car for you.
These systems use a range of sensors including cameras, infrared and LIDAR along with extensive maps databases to drive without human intervention. Alphabet, the parent of Google, is the company that is furthest along in fully autonomous vehicles.
In Google’s testing, there have been no fatal accidents. The only accident caused by a Google vehicle was a very-low speed collision with no injuries.
A long transition
We are in the midst of a long transition. Unfortunately accidents will happen because of a combination of human laziness, overselling of the product and confusing interfaces. The current semi-automation systems have a lot of limitations.
I recently rented a Cadillac STS with a lot of these features. As I drove it, I tried using the “lane keep assist” feature. In theory, the system would keep me in my lane. I tried it on curvy Interstate 280 in the Bay Area, in moderate traffic. As far as I can tell, the system didn’t work. When I took my hands off the wheel, the car would drift a foot into the other lane before pulling me back into my lane. Although I’m a big fan of testing products to the limit, I wasn’t about to do that in traffic.
It’s possible that it was user error. Or a confusing interface. Or I was outside the limitations of the system.
According to GM, Lane Keep Assist and Lane Departure Warning systems may not:
- Provide an alert or enough steering assist to avoid a lane departure or crash
- Detect lane markings under poor weather or visibility conditions, or if the windshield or headlamps are blocked by dirt, snow, or ice, if they are not in proper condition, or if the sun shines directly into the camera.
- Detect road edges
- Detect lanes on winding or hilly roads
And if Lane Keep Assist only detects lane markings on one side of the road, it will only assist or provide a Lane Departure Warning alert when approaching the lane on the side where it has detected a lane marking.
Lastly, GM says that using Lane Keep Assist while towing a trailer or on slippery roads could cause loss of control of the vehicle and a crash. Turn the system off.
When the LKA or LDW systems don’t work properly, the system performance may be affected by:
- A close vehicle ahead
- Sudden lighting changes, such as when driving through tunnels or direct sunlight on the camera sensor
- Banked roads
- Roads with poor lane markings, such as two-lane roads
That is a lot of limitations to be aware of! It’s too easy to learn to rely on semi-autonomous features that might work 95% of the time but have dire consequences in the 5% case.
Marketing doesn’t help either. The benefits are highlighted in glamorous videos; the limitations buried in fine print. Even naming makes a big difference. Calling something “Autopilot” given the state of today’s technology is vastly overstating the case.
Car companies aren’t the greatest at user-interface design, often using what look like hieroglyphics for controls. In my test of the STS, I thought the car had an automatic braking system based on the icons. I’m glad I didn’t try to test that — because it didn’t. Mine was a somewhat unfair test because if I owned the vehicle, I’d probably know what features I had. But if someone had been borrowing my car, they’d be presented with the same set of challenges.
Driver training on the proper use of new features is key. When I went through driver’s ed, I was taught to pulse the brakes to prevent the wheels from locking up. But with antilock brakes, you are supposed to step hard on the brakes. I was taught to put my hands at 10 and 2 on the steering wheel; with airbags, you want to put them at 5 and 7.
Not only are the controls of new features not intuitive, some companies even fiddle with basic features.
FCA’s redesign of the transmission shifter is mind-bogglingly stupid.
The National Highway Traffic Safety Administration’s investigation into the Monostable gear shifter used by a number of Chrysler, Dodge and Jeep vehicles is turning into a recall; FCA will recall approximately 1.1 million vehicles worldwide to modify the operation of the shifter that has now caused 121 accidents and 41 injuries.
The issue itself is not a fault of engineering but rather design, as the shifter returns to the default center position without giving the driver sufficient feedback as to the selected gear.
As a result, a number of owners have exited their vehicles thinking that they had put the vehicle into Park, while in reality it remained in Drive or Reverse position. The NHTSA has called the operation of the shifter “unintuitive” and had opened an investigation into the issue months ago.
With driver reliance on semi-automation systems, system limitations and confusing user interfaces, we can expect to see more cases like the Tesla accident.
Media frenzy and public irrationality
My big worry is that media hype around the small number of accidents will hurt the development of truly autonomous vehicles that can save a lot of lives.
Even in the current state, semi-automation features like lane management and automatic braking can save lives. IF drivers use them as backups.
But we’ll see endless stories about how dangerous automation is. Anything that is “new” is dangerous. It was worldwide news when a Tesla caught fire. Never mind that gas vehicles catch fire much more frequently.
Imagine if we had 24-7 news networks during the rise of aircraft. In the early years of aviation, lots of accidents happened. Every accident would have been covered nonstop.
With much less media scrutiny than we have today, we were able to improve airliner safety. With every accident, we investigated, learned what went wrong and improved.
The NTSB is great at what it does. Although we primarily hear about it in the context of airline accidents, they’re already looking into the Tesla accident.
They provide reasoned analysis, tradeoffs and recommendations. Unfortunately, government, politicians, media and the public don’t work that way. We will see negative hype around self-driving cars as politicians chase votes and media chase ratings.
When it comes to media, only the misses count. If your technology saves 4,999 people, you don’t get credit for that. But you get dinged for the one it doesn’t save.
Developing our safer future requires some reasonableness on the part of consumers, manufacturers, media, politicians, regulators and attorneys. Is that an unreasonable ask?
We know what self-driving cars look like today. We’ve seen two models so far: the Lexus SUV with a bunch of cameras and sensors on it and the Google bubble car.
But what might they look like 35 years from now? Here are some thoughts.
The obvious design changes will be removing the steering wheel, dashboard and gear shift. But those are some of the least interesting.
Cars today are designed for human drivers, safety and convenience. Over time, the first two become unnecessary and the convenience becomes dominant.
There are a number of features that are necessary for driving that can go away:
- Driver and passenger side mirrors and the rear view mirror. Nothing to see here. And getting rid of the exterior mirrors will make the cars more fuel efficient.
- Windshield wipers. The car doesn’t need these to see. (But there are other reasons to keep them — see below.)
- Windows. The car doesn’t need them, but also see below.
- Turn signals.
- Dashboard with gauges.
Some things can’t go away, not because the car or passengers need them, but for safety of others. We don’t really need headlights. But pedestrians do. However, the headlights can remain off most of the time and be turned on when there are people or animals around. This will dramatically reduce light pollution. (Lack of signals have already caused problems for humans; the blind can have a hard time with hybrid and electric vehicles because they are nearly silent.)
We’ve added a lot of safety features over the years:
- Side impact door beams
- LATCH anchors for car seats
- Crumple zones
These add substantial weight to cars.
With self-driving cars, car accidents will be incredibly rare events. The biggest safety feature will be the lack of human drivers.
Make driving safe enough and you can get rid of that weight, making cars more fuel efficient.
Sure, we’ve added radios, CD players, rear-seat DVD players and cupholders. But a self-driving cars will provide a whole new dimension of convenience.
Instead of having the typical front-facing seats, we can have different seating arrangements. Maybe a table for playing games, cards or just talking.
Recliner seats or beds for sleeping. Reading lights that dim the rest of the cabin.
Air travel is a model to look at: from the perspective of the passenger, an airplane is a self-driving transportation vehicle. We could have a big screen display for watching movies with thousands of options. Cameras for teleconferencing. Better, more immersive sound systems.
We could have the equivalent of Airshow: maps and stats on the journey.
These features could be segmented as airplanes are. Vehicles that are basic for short trips and luxury vehicles for long hauls. (Of course, you could order these on demand for a particular trip.)
There are some things that the car doesn’t need, but we might want to keep for humans. Windows and windshield wipers are two of them.
We still have windows on planes because people want to be able to look out. (Cargo planes don’t have windows because it is more fuel efficient.) Likewise, we need to provide visibility, especially on scenic roads. But we can improve these, too: windows and the windshield will have the ability to become opaque. This is better for having sex and watching movies. Or driving on an urban blight road full of billboards.
Now all of this will take a long time. The average vehicle on the road today in the United States is more than 11 years old. If we’re looking at individually owned vehicles, it would take 20 years or more to turnover the fleet. But this should be accelerated by purchase of self-driving cars by companies like Uber and Lyft for on-demand service.
The driving and convenience features are easier to change than the safety features.
Having a self-driving car on the road without a steering wheel is fine, because other vehicles don’t rely on it. We can’t get rid of turn signals because other cars need to know which way the self-driving car intends to go.
The safety features will need the longest to get rid of. If humans can cause accidents, we still need to protect the occupants of the self-driving car. This also affects the convenience features; we can’t have people standing up unbuckled if there’s a chance that the car will get hit.
Regulators and public fear will also play a role in further delaying the removal of outdated safety features.
All of these changes can’t happen fast enough for me.
Read my post on how cars have changed over the past 35 years.
Last night’s episode of Silicon Valley is one of the must-watch episodes of series.
Our friends at Pied Piper have accomplished an amazing technological feat. Their friends and family (except for Monica) love the product. Downloads are going crazy. We watch along as the counter reaches 500,000 downloads. But look under the hood and there are fewer than 20,000 daily users.
Pied Piper commissions a focus group where it becomes clear that consumers don’t understand the product. Richard spends hours trying to explain it to them.
What went wrong?
If you’re designing a product for the masses, it needs to sell itself. The benefits (not the features) have to be obvious quickly. Marketing like this isn’t the answer:
(If you hire an agency — which you shouldn’t do initially– and they create something like this, fire them.)
Pied Piper’s platform was a product by engineers for engineers. It was tested by an initial beta group that was mostly engineers.
Here’s the Pied Piper user interface:
This reminded me a lot of product I worked on. Our CTO constantly wanted to add new features to the platform. As a result, the site looked a lot like this.
Our platform had more (and better and cheaper) features than the competitors. But we were putting all of them in front of the consumer at once. People couldn’t understand what they could do with it.
We broke the product up into three products that solved different consumer needs. (Benefits instead of features.) This was mostly a design and marketing effort, but obviously the the engineers had to build it. All of the products ran on the platform we’d already built. (It was just different skins that adapted the platform to specific use cases.) We also raised prices closer to our competitors. Despite the price increase, demand went up. Once people understood what we were selling, they were willing to buy.
Real-life examples abound. Wave and Google+ are two of the highest profile examples. Googlers tell me that internal testing was off the charts. (A more detailed post on why Google+ failed coming soon.)
If you want to design great consumer products, you need to have an interdisciplinary approach. You want a designer who has worked on complex consumer-facing products. You need a marketing person who has knowledge of consumer behavior. They all need to be working together on the product. These aren’t necessarily separate people; often, one person can wear multiple hats.
One of my recommendations: send your product to someone you know is not technical, maybe a parent or sibling. Give them a task list. See how they do. (Ideally, with screen sharing.) It’s important to do it without guidance because no one will be guiding them in real life.
When I tweeted about this, I got this response:
This sounds great — solve a problem that people have. But the greatest value comes from solving problems people didn’t know they had. The iPhone and Facebook are great examples.
What existing problem did iPhone solve? It created a completely new category that people fell in love with.
Facebook is similar.
It’s easy, in retrospect, to define problems that were solved. In Facebook’s case, you can easily keep in touch with tangential contacts, such as high school and college classmates, business acquaintances, etc. Sure, there is clear demand for this now. But I don’t know that there was an unrequited need to share pictures of your lunch with your friend from 3rd grade.
Some of the key reasons for Facebook’s success:
- Simplicity. The initial feature set was very limited and easy to understand. Even if you have engineered 4,000 features behind the scenes, the initial experiences should be easy-to-get. You can expose some of the other features later.
- Iteration. Facebook rolled out market-by-market (Initially, Harvard and then other elite schools.) Only later did it expand to the masses. By the time it was rolled out, behaviors had been established. (Poking, status updates.) It’s easier for people to mimic behavior of others than to create their own behaviors.
- Growth and marketing built-in. Facebook is great at this; Google sucks at this. The way products succeed today is by having growth mechanisms built into the product. See my post on how people tagging was key to Facebook’s success.
When you’re designing products for consumers, there is no such thing as too simple.
Today’s the 5-year anniversary of the launch of Google+. It was an unmitigated disaster for Google. Despite spending many man-years of development, endless hype in the media and Google’s attempt to cook the books on usage stats, the network is essentially dead.
Google+ failed for a simple reason: It blatantly tried to copy Facebook instead of playing to Google’s strengths.
We’ve seen a lot of attempts to copy successful products of others. Facebook tried to compete in search. Facebook tried to copy Flipboard (Paper), Instagram (Camera) and Snapchat (Poke). All of these attempts failed.
The only product in recent memory where the copy was more successful is Facebook Live, which is essentially Meerkat. I’d argue this was because Meerkat didn’t really solve a compelling user problem. Most people don’t need to broadcast 1-way video. Those that do need broad distribution, which Meerkat lost as soon as it was cut off from Twitter. (To the extent people want video, it’s 2-way, such as FaceTime, Skype or Hangouts.)
The reason these copies didn’t succeed? They didn’t incorporate what was unique about the new platform; what made them successful. In Google, that is search. In Facebook’s case, that’s social.
Google+ required you to replicate what you’d already done on Facebook. Create a profile, friend people and post. The unique and much better features of Google+ — Hangouts and Photos — were buried by comparison to the Facebook- product. Why would anyone repeat all the work they were doing on Facebook on Google+? Or switch to a platform where none of their friends are for no real benefit?
Google embedded Google+ everywhere it possibly could (YouTube comments, giant alerts, etc.) But it didn’t effectively do it where it mattered: in search. Hundreds of my friends use Google everyday. The results that they click on are more likely to matter to me than results that the general population click on. Despite the fact that I have a network of hundreds of people, I’m still searching in isolation.
If my buddy Bob spent 2 hours researching a trip to Senegal, shouldn’t I be able to learn from his efforts? Shouldn’t I be flagged that Bob did this work, maybe went to Senegal and had knowledge on the topic? Maybe I should reach out to him and learn about it? (Of course, this always needs appropriate privacy permissions. I shouldn’t be able to see Bob’s searches unless he makes them available to me.)
A friend wrote a review in Google’s local product of Rosewood Sand Hill. That should be front-and-center on this screen. It’s what I would consider by far the most relevant. But it’s nowhere to be found.
The right way for Google to play in social is to add a social layer to Google. If the value proposition to the consumer was “have your friends help you search,” instead of “use a version of Facebook without your friends,” I imagine Google+ would have been much more successful.
People search on Facebook. All the time.
Conversely, most of Facebook’s efforts on search, have focused on the search box. People search on Facebook all the time. But they don’t search in the search box, they search in status field.
If Facebook copies Google’s definition of search, they will (and have) failed.
What do I mean by people search on Facebook? Consider this example:
This is no different than a Google search for “Senegal”. Except, I am asking my friends, in a highly inefficient manner. There’s a high likelihood that someone in my friend network (of 600+ people) has been to Senegal or knows something about Senegal. But my post doesn’t efficiently reach those people. FB, through, NLP should identify this as a query for “Senegal” and present this post to my friends who have been to Senegal.
That creates a better search experience because I get expertise from people I actually trust.
If you expand distribution to friends of friends, you are almost guaranteed to find someone who has an answer. In this case, in an efficient way, my friend Mandy has expanded the search to her friend Chris in the last comment.
It could either be highly prioritized in news feed for them, or they could get a notification that says “Your friend Rakesh is looking for information about Senegal? Want to help him out?”
Modifying Facebook in this way also helps improve the social experience and increases the liquidity in the market. By expanding the distribution to my friends most likely to know the answer, I get an answer faster. This also opens up the possibility of creating new relationships or renewing old ones.
- I haven’t talked to friend Bill in a while.
- I post a “query” for Senegal.
- FB knows that Bill has been to Senegal. (Pictures posted from there, status updates from there, logins from there, etc.)
- FB surfaces the “query” to Bill.
- Bill sees it and responds.
- Bill and I reconnect.
Fact-based queries vs. taste-basted queries
This all works better for matters of taste vs. fact. Google is going to give you a much better, quicker answer for queries like the “value of pi” or “5+2” or “weather in Miami”.
Yes, I could ask this in Facebook — and I did:
More than an hour later, I still had no answer. (And my non-technical friends, who didn’t know what I was doing, would think I’m an idiot.) Mihir asked about chatbots — I’ll get to this in a minute.
But those are matter of facts — and, btw, have zero advertising against them.
Think about queries like “plumber,” “dentist,” “lawyer,” “auto insurance”. Those are queries of taste. And, it may shock people, but that’s where you make your money in search! Travel, law, professional services and insurance are among Google’s top money makers.
While many people, including Wall Street analysts, treat search as a monolith, search is actually a collection of verticals. Each has different levels of monetization. Many fact-based queries have no advertising against them.
Facebook doesn’t have to solve the queries of fact. Leave those to Google. (It could, but people aren’t searching FB for those.)
Facebook can pick off the higher-value queries and the ones that are most likely to add to the FB experience and value proposition: a place where you come to interact with your friends.
FB can also use these “queries” as a way to turn its ad into higher revenue, intent-based ads. In addition to your friends comments, you’d see — clearly identified — responses from advertisers to your query.
Someone who posts a query “anyone know of a good hotel in London?” could be presented with an advertiser comment for “hotels in London.” This presents a highly relevant ad that someone could turn to immediately. (It could also be time delayed — if I don’t get a response from a friend, the advertiser comment shows up.)
Facebook is trying to do this in a ham-fisted — and annoying and needlessly interruptive way.
I was recently hit by an Uber while walking across the street. My cousin asked me about it on Messenger. Here’s what happened:
My cousin is asking how I’m doing after I was hit by an Uber. Messenger is throwing an ad for Uber in both of our faces. (Not only once, but three times. See my post on bots.) There are some great uses for bots. Sticking irrelevant ads in front of people isn’t one. (I’ll talk about good use cases in a future post.)
Often, you’re forced into a space by business needs or the stock market demanding that you have a “search” or “social” strategy. Or there’s a hole in you business model. See also: wireless carriers in payments, video, content, pictures.
The easiest thing to do is to try to copy someone else who has been successful. But if they’re already dominant, how are you going to win? You can’t just create something to plug a hole in your business strategy; you need to plug a hole in the customer’s needs.
These are just two big examples of how you could win by playing to your own strengths — and your user’s frame of reference about your product.
When designing new products, you should figure out what makes you different and better. Then build off that.
The Agrawal family cars over the years. The Pontiac Catalina was our family’s first car when we immigrated to the United States. The Mini Cooper is my current car. (These are images gleaned from Google search, not my actual cars.)
What will cars look like 30 years from now? That’s a question I’ve been talking to friends about recently.
One way to assess the future is to look back at the past. My memories go back to when my parents drove us 35 years ago.
Safety features: shoulder belts, airbags, antilock brakes, crumple zones, electronic traction control, reinforced door beams, LATCH anchors. A passenger side mirror was an option; these days it’s standard. Third brake light. Xenon headlights. Tire pressure monitoring system. Some cars have wipers for lights. (Relatively rare.) LED brake lights. Back up sensors and back up cameras. High end cars can have night vision, pedestrian protection. Many of these changes are because of regulations, but they started because of new product development. For example, airbags were in some cars before they became mandatory. Backup cameras aren’t required today, but they’re in a lot of vehicles. But in 2018, they will be required.
Convenience features. Power locks/windows, air conditioning have become standard on all but the cheapest cars. Remote to lock/unlock. Reading lights, vanity lights. Rear defogger. Mid-level cars with keyless entry, power mirrors, power seats, auto dimming rear view mirror, heated seats, push button start. Garage door opener. Rain sensor. Automatic climate control, dual-zone climate control. Memory seats/mirrors. Voice recognition. Carplay, Android Auto, High end cars add self-opening glove box (really!), heated steering wheel, seat coolers, a zillion seating controls, massaging seats, electronic air freshener (really!)
Automation. Only 4% of cars sold today in US are stick shift. (vs. 31% 30 years ago.) Auto stick, paddle shifters. Cruise control, intermittent wipers. Power steering, power brakes. High-end cars have adaptive cruise control, blind spot warning, lane departure warning systems, automatic lane changing, automatic braking, self parking.
Infotainment/telematics. Tape decks have gone away, CD players replaced them and on Teslas, there isn’t even a CD player. Nav systems available on mid-high end cars. Minivans for a while had DVD players built into roof for kids. Bluetooth. Not widely adopted, but available: satellite radio, OnStar. Some cars can be managed by an app.
Infotainment/telematics. Tape decks have gone away, CD players replaced them and on Teslas, there isn’t even a CD player. Nav systems available on mid-high end cars. Minivans for a while had DVD players built into roof for kids. Bluetooth. Not widely adopted, but available: satellite radio, OnStar. Some cars can be managed by an app.
Repairability. Cars have become more difficult for the weekend mechanic to repair. Some things are still easy, like oil changes and filter changes. But a lot of the car is controlled by complex electronics instead of mechanical parts.
Software and connectivity. Cars are increasingly run by software. This can create problems when cars are poorly designed; recent Jeeps could be remotely controlled because hackers went to the CAN-bus through the infotainment system. Software can be upgraded. Some Fords can be upgraded through the USB port. Teslas update wirelessly and the company can add significant new features. The industry has standardized on the OBD-II connector. That sounds like it’s not important — but it’s provided a pathway for innovation. For example, my car insurance is charged per mile and I can get speed information via a dongle that plugs into the OBD-II port. The dongle also provides diagnostic data that I can read on my insurance app.
Pollution. Our first car ran on leaded gasoline. Now, most cars run on unleaded gasoline. Unless they’re running on electric, which has zero-emissions. On gas-powered engines, CO2 emissions are a fraction of what they were 30 years ago. Fuel economy has nearly doubled from 13.1 MPG to 24.7 MPG.
Design. New form factors like SUV and minivan. Station wagons rare. Color keyed bumpers. (I miss chrome!) Hood ornaments have disappeared in place of badges on the front of the car. Antenna design has become increasing sophisticated and the antennas have become less visible. Vinyl tops have gone away. (Thankfully!) Cars have become much more aerodynamic.
Etc. More/better diagnostics. Instead of simply dummy lights, many mid+ cars have engine computers. Digital speedometers instead of analog. (Some cars.) Outside temperature sensor. Vinyl to cloth seats. Bench seats in front replaced with bucket seats. Fold down seats with trunk access. Cigarette lighters are less common. Even on my ’95 Nissan Altima, the lighter element was missing. (A 12V power plug was there, but the lighter was an accessory.) With the decline of smoking, ashtrays have become less common. We see USB ports, auxiliary inputs and iPhone connecters.
And, of course, cup holders!
It’s important to note that this is based on the U.S. market. Despite globalization of the auto industry, there is significant variation from market-to-market. Manual transmissions are very common in Europe and Asia; they’re rare in the U.S. Emissions standards are different. There is also right drive vs. left drive.
So what does this tell us about the future? It says cars evolve slowly.
But I expect that the pace of innovation to speed up for a number of reasons:
- Increasing competition from other car manufactures. The “big three” aren’t as dominant as they used to be. Chrysler is merely a brand. GM has shuttered Oldsmobile, Pontiac. Saturn was created and shuttered in that time frame.
- Cars are now designed using software, making it easy to test variations before going through the expense of tooling.
- Standardized parts suppliers. In the old days, car manufacturers had captive parts arms. Now companies like Continental, Alpine and Johnson Controls provide many of the parts. This allows innovation to spread faster. To some extent, car companies are systems integrators.
Of course, the biggest push toward automation is being driven by Google with its focus on automation.
But there are also a lot of factors that slow innovation. Despite the fact that technology can improve quickly, the regulatory environment doesn’t because of the slow pace of legislative change.
The basic process of buying a car has been virtually unchanged. You go into a dealership and haggle with a guy (still mostly guys) who then “talks to his manager” to “see what he can do” to “put you in that car today.”
Behind the scenes, many car dealers are really owned by the same companies despite having localized brands. Heard of Penske Automotive? Probably in the context of auto racing. Penske owns 243 dealerships generating $15B in annual revenue. Single rooftop dealers are much rarer.
With today’s market, the right way to sell cars is direct-to-consumer by manufacturers. But dealers control state legislatures with promises of campaign contributions and local jobs. In some states, you have to go through machinations to buy a Tesla.
Politics plays a role in other ways. In Florida, a candidate used an ad to attack an opponent for supporting self-driving cars. Scaring the shit out of seniors is a time-honored practice. Never mind that self-driving cars would give seniors with vision problems, reduced reaction times and limited mobility a lot more freedom.
Some regulation is done on a state-by-state basis. Legally, a car can’t drive itself across the country.
There is much work to be done. In the road ahead, I’ll look at the effects of automation on the design of cars themselves and the effects on society.
I’ve always hated job interviews. On both sides. Not only are they poor indicators of eventual success, they also create a dynamic that isn’t good.
Some of the things I hate:
- They don’t allow for the possibility that the interviewee is smarter or has a level of experience that interviewer doesn’t have. This is especially true when you have a marketing person interviewing an engineer. Often, the only assessment that can be made is cultural fit.
- There’s generally no way to assess the interviewer. There have been several cases in my professional life when I know the interviewer was a terrible interviewer and not getting any insight. In some cases, an asshole on the interview loop may by annoying prospective candidates to the point that they don’t want to join the company. There should be a mechanism for an interviewee to rate an interviewer. (The incentives are complicated here, but I can think of some ways.)
- The process rewards people who know the tricks of interviewing. Because it’s only 20-25 minutes per interviewer, it is often easy to blow through the process with prepackaged talking points.
The biggest issue is that they create a confrontational dynamic, instead of a conversational dynamic.
Here are some of the questions I use when interviewing marketing and product people. In most cases, there is no “right” answer. I’ve often learned something when talking to interviewers. But for some of them, there is a definite wrong answer.
- Late-night talk show host Jimmy Kimmel comes up to you with a camera crew and asks you, “Who is the president of the United States?” What should you say? Why?
- A pizzeria charges $12 for a 9″ pizza. How much should it charge for an 18″ pizza?
- Financial analyst frequently beat up on Google because its CPC is declining (the revenue generated per click). Is a declining CPC really bad for business? Why or why not?
- There are extremely rare circumstances where a self-driving car will have to chose among hitting two people. How do you decide which person to hit?
- In question 3, what if the pedestrians could be identified as Stephen Hawking and a Wal-Mart clerk?
- In the self-driving car scenario, assume the car’s options are: hit a deer head on or swerve into oncoming traffic. If you hit the deer, there is a high probability that the driver will die. If you swerve into oncoming traffic, there is a lesser probability that you will die. But you create a risk for the other driver that he could be injured or killed.
- You accidentally get a Vanguard statement that was supposed to go to a well-known psychic. Does this make you believe in her skills more or less? (For the purpose of this question, assume you have some level of belief. You can’t opt out of the question by saying psychics separate the gullible from their money.)
- In the heart of Time Square, there has been a tkts booth since 1973. The booth offers 1/2 off tickets many Broadway shows. People line up and wait for an hour or more to get these cheaper tickets. Obviously, technology has changed a lot since 1973. They have a great app whose functionality could be enhanced for online ticketing. The 1 hour wait would go away and offer theatergoers instant access. Should they add app based ticketing? Why or why not?
- You are product manager for an OnStar like service. The capabilities include remote door unlock, vehicle status reports, turn-by-turn directions for navigation (talking to an agent to enter the destination), warnings when you need to go the dealer for service. The technical capability is in every car. The package costs $200 a year for unlimited use of all services. It’s possible to offer a one-time remote unlock service. It would cost you $1 to do a one-time unlock. It happens instantly. The consumer’s other alternative is to call a locksmith. The locksmith has to pay a technician $40. The retail price is $80. Again, assuming it only costs you $1 to provide the service, should you offer the a la carte product? If so, how much should it cost?
- Back to Question 1. Does your answer change if you’re a 24-year-old aspiring actress? If so, why?
- How would you market Twitter?
Bots are all the rage in Silicon Valley these days. Everyone is talking about them. Facebook has a bot built into Messenger. Google announced a platform for bots. Many startup pitches focus on bots.
But, so far, I’m not convinced.
I’ve worked on bot-related projects. I’ve worked in messaging, speech recognition and natural-language processing. At AOL, we had a local search bot in 2004. (Not my project, but we had one.)
Today’s bots are pretty stupid. They are based on simple keyword recognition. See this example from a recent Messenger conversation I had after I was hit by an Uber:
Ideally, the bot would realize the context and not even show the “Request a Ride” message. At a minimum, it should be frequency capped so it doesn’t appear every message.
Like I said, bots today are stupid.
Will they get better? Undoubtedly.
Will they replace a personal assistant in the near future? Unlikely.
Consider an obvious use case for bots: appointment scheduling. After all this time, appointments are still tough to schedule.
It sounds simple. Check two calendars and find overlapping free spaces. But it’s not.
Availability isn’t fixed. My availability can vary based on who is asking. Mark Zuckerberg wants a meeting? My calendar is wide open.
For me, multiple factors go into availability. What is my relationship to the person? Are either of us working on an important project? Have I met this person recently? Were they referred by someone I trust?
Have the assistant look at my address book, you say? Well, relationships vary over time.
Location matters. Again, this is context sensitive. I one joked with VC Shervin Pishevar that location varies based on importance of the two parites. Jokingly, we scheduled lunch at Madera, which is basically across the street from most venture firms. But there’s a lot of truth to that. If I’m meeting a friend who lives on the peninsula, we try to alternate peninsula and SF. If I’m meeting a VC, it is usually in Menlo Park. Meeting someone who needs my advice? Probably a coffee shop a few blocks from me.
Location also depends based on our local travel patterns. I try to schedule my meetings on the peninsula on one day to minimize my driving. I know some people spend one day in SF. I adjust to take that into account.
The rules vary for out-of-town visitors. Some of my friends visit from the East Coast. Some do it every other week. I can see them pretty much any time. But for a friend who visits once every six months. I’ll work hard to open up space.
Location can be a Skype call if I need to chat with someone who I won’t likely meet up with.
Timing matters. Is the request urgent? I’ll open up availability. If not, it might wait a couple of weeks.
I’m having surgery next week. It was scheduled according to the surgeon’s schedule because she has a specialized skill. I scheduled everything around it. During the recovery period, I’m not taking meetings. But if someone and something important needs to happen, I’ll suck up the pain.
Then there’s the issue of social projection. Maybe I don’t want to project that I have nothing going on and my calendar is wide open. Or I want to project that someone is really important to me.
Something that seems to be so simple can actually be really complicated.
Is my scheduling more complicated than most? Probably. But if it weren’t complicated, the utility of a bot would be less valuable. Could a bot learn all of these rules? Maybe.
It’ll take a lot of time to get there.
Maybe a bot can put it in on my calendar for 2020.
On a recent trip to New York City, I did some secret shopping of banks. I posed as a customer looking to open a new account with $250,000. I visited banks large and small, including Citibank, Bank of America, Chase, Valley National Bank and TD Bank.
Although banks have become more open and inviting to consumers — in many neighborhoods, it isn’t a prison environment with tellers behind bulletproof glass — I came up with a number of ways to update the banking experience. Some of these require changes to approaches to security, but I think they are necessary for banks to be relevant.
- Ability for bank staff to access Dropbox, email, etc. There are obvious security implications of this, but this is the way millennials work. The primary repository for many documents are in cloud storage. If your bankers don’t have the ability to access Web services, you’re at a disadvantage. I love that my banker at First Republic can do most things over email and Dropbox.
- Web access for certain employees. Although some bank networks are locked down for the lowest common denominator (bank tellers), some people need access to the broader Web. This includes investment advisers and personal bankers.
- Simpler ATMs that allow most of the transaction to be done from the customer’s mobile device.
- Allow customers to enter their data. Although many banks use PIN pads, there is other data that needs to be entered. iPads could allow people to comfortably enter data such as email addresses.
- Bathrooms! In the past, bank transactions were predominantly about quick withdrawals and deposits. Now, more transactions require more time. This includes account opening and investment transactions. When you might be in the bank for 15-20 minutes, bathrooms become more important. In my secret shopping, most banks wouldn’t offer a bathroom. A few did, but it involved going into a backroom.
- Chargers. Just like bathrooms, charging devices are becoming a Big Necessity. In some of my secret shopping, bankers allowed me to use their personal chargers or asked around the bank for one. It’s not expensive to outfit each station with micro USB and lightning charges.
- Better IVR and ASR systems. One of the more frustrating banking experiences is calling a bank, entering the account number and then having to give it again to the rep. These systems need a streamlined user experience with user needs in mind. They shouldn’t be one-size-fits-all. For example, if a bank notices that I do all my banking online, I should go straight to a representative. (The chances that I will call for something I can do online are zero.) People who never bank online would be given the full automated system. Segmentation could also be done by asset level, e.g. customers with more than $100,000 go straight to a rep.
- Allow CSRs to follow transactions. For some transactions and events, there is a time delay between a request and a final outcome. Normally, this involves the customer calling back and dealing with someone brand new and re-explaining the situation. Citi lets an agent monitor an account for future follow up. I’m always impressed when the same person calls me back with a resolution.
- Automatic emails for frequent transactions. For example, if a customer requests a late-fee credit, an email is automatically generated stating the amount of the late fee credit and an expected post date. Ideally, these emails will show up before the customer hangs up.
- Data should be shared across platforms. If I start a credit card application online, I should be able to finish on the phone if I have a question.
One thing I noticed that really impressed me: Chase bankers rate other Chase employees. My banker had to call a support line. After he finished, he was presented with a survey to rate the support desk. This kind of feedback is invaluable in weeding out bad employees.